Who would have though including a hard depedency on third part service with unclear long term availability would be a problem!
Paid compilers and remotely acessible mainframes all over again - people apparently never learn.
Paid compilers and remotely acessible mainframes all over again - people apparently never learn.
The destruction of spelling didn’t feel like doomsday to us. In fact, I think most people treated the utter annihilation of that skill as a joke. “No one knows how to spell anymore” - haha, funny, isn’t technology cute? Not really. We’ve gone up an order of magnitude, and not paying attention to how programming is on the chopping block is going to screw a lot of people out of that skill.
Zuckerberg recently said that those not using AI glasses will be cognitively disadvantaged in the future.
Imagine an AI-native from the future and one of the fancy espresso coffee machines that we have today. They will be able to know right away how to operate them from their AI assistants, but they won't be able to figure out how they work on their own.
That's the future that Zuckerberg wants. Naturally, fancy IT offices will likely be gone. The AI-native would have bought the coffee machine for nostalgia effect for a large sum, trying to combat existential dread and feeling of failure which are fueled by their behavior being even more directly coerced into consumption.
Sure, calculators are useful in professional life and in high school math and sciences. But you still had to do everyday math in all kinds of places and didn't always have a calculator at hand. The smartphone changed that
I feel that's relevant in two ways: just like with math, a little bit of manual coding is going to be a huge difference compared to no manual coding, and any study like the one you propose would be hugely complicated by everything else that happened around the time, both because of smart phones and the coinciding 2008 crash
For developers who read and understand the code being generated, the tool could go away and it would only slow you down, not block progress.
And even if you don’t, it really isn’t a hard dependency on a particular tool. There are multiple competing tools and models to choose from, so if you can’t make progress with one, switch to another. There isn’t much lock-in to any specific tool.
I think there is a lot of reticence to adopt AI for coding but I'm seeing it as a step change for coding the way powerful calculators/workstation computers were for traditional engineering disciplines. The volume of work they were limited to using a slide rule was much lower than now with a computer.
Recent research suggests it would in fact speed you up.
https://metr.org/blog/2025-07-10-early-2025-ai-experienced-o...
So even if it’s not an actual productivity booster on individual tasks, perhaps it still could reduce cognitive load and possibly burnout in the long term.
Either way, it’s a tool that devs should feel free to use or not use according to their preferences.
I don't think this one is a good comparison.
Once you had the binary, the compiler worked forever[1]
The issue with them was around long term support for bugs and upgrade path as the language evolved.
---
[1] as long you had a machine capable of running/emulating the instruction set for the binary.
Clang is not being professionally used commonly in the embedded space.
IAR for example is simply a fantastic compiler. It produces more compact binaries that use less memory than GCC, with lots and lots of hardware support and noticeably better debugging. Many companies have systems-engineering deadlines which are much less amenable to beta quality software, fewer software engineering resources to deal with GCC or build-chain quirks (often, overworked EEs writing firmware), and also a strong desire due to BOM cost to use cheaper/less dense parts. And if there is a compiler bug or quirk, there is someone on the other end of the line who will actually pick up the phone when you call.
That said, some of those toolchain+IDE combos absolutely do suck in the embedded world, mostly the vendor-provided ones (makes sense, silicon manufacturers usually aren't very good at or care much about software, as it turns out).
That's true in general. With paid licenses and especially subscriptions, you're not just getting the service, you're also entering a relationship with the provider.
For companies, that often matters more than the service itself - especially when support is part of this relationship. That's one of many reasons companies like subscriptions.
For individuals, that sucks. They don't need or want another relationship with some random party, that they now need to keep track of. The relationship has so much power imbalance that it doesn't benefit the individual at all - in fact, for most businesses, such customer is nothing more than a row in an analytics database - or less, if GDPR applies.
(Also AVR/PIC compiler from Microchip had a dongle as recently as February this year, and it looks like it's still available for sale even though its license isn't in the new licensing model).
What's the cost between never using Claude, and using it and getting these lower limits? In the end, I'm left with no Claude in both situations, which leaves me better off for having used it (I wrote more code when Claude worked).
As for the cost, you are ignoring the situation where someone has depended on the tool so much that when it goes away they are atrophied and unable to continue as before. So no, in your scenario you’re not always left better off.
They don't, because ironically enough human meatbags all need to be trained from scratch, unlike LLMs which don't die every 80 years or so.
I doubt even 10% of the devs now were conscious when "paid compilers and remotely accessible mainframes" still existed.